BEHAVIOR Challenge 2026 β€” task 11 (putting_dishes_away_after_cleaning), low-lr fine-tune (skill stages), step 2999

Selected checkpoint for the BEHAVIOR-1K 2026 submission.

Model

  • Architecture: the 2025 BEHAVIOR Challenge winner's pi0.5-based policy (IliaLarchenko/behavior-1k-solution, ~3B params, task-ID embeddings, no language prompt).
  • Init: the winner's task-specific checkpoint_1, fine-tuned (not used unchanged).
  • Training config: pi_behavior_b1k_2026_task11_lowlr (in our pipeline repo's behavior-1k-solution/src/b1k/training/config.py), with skill-annotation subtask stages.
  • Data: 2026 challenge demos, task 11 episodes 2200-2379 (180 demos), RGB cameras only.
  • Recipe: 3000 steps from the winner's checkpoint_1, batch 16, cosine lr 5e-6 -> 1e-6 (warmup 100), skill stages. This is the final step 2999 (no screening; straight to the official run).
  • Checkpoint is params-only (no optimizer state): it can be served or used as an init, not --resumed.

Layout

  • params/ β€” orbax/ocdbt parameters
  • assets/ β€” normalization stats
  • _CHECKPOINT_METADATA

Serve with scripts/serve_b1k.py --task-id 11 policy:checkpoint --policy.config pi_behavior_b1k_2026_task11_lowlr --policy.dir <this dir>.

Evaluation (q = fraction of goal predicates newly satisfied)

Scored the 2026 leaderboard way: public instances 0-9 (ids 301-310), one rollout each, default 1.5x timeouts, task score = sum / 10.

Leaderboard score: q = 0.0929 (sum 0.9286 over 301-310; full completion: none).

instance 301 302 303 304 305 306 307 308 309 310
q 0 0 0 0 0 0 .429 .500 0 0
Downloads last month

-

Downloads are not tracked for this model. How to track
Video Preview
loading